692
Chem. Res. Toxicol. 2002, 15, 692-698
Arsenate Reductase II. Purine Nucleoside Phosphorylase in the Presence of Dihydrolipoic Acid Is a Route for Reduction of Arsenate to Arsenite in Mammalian Systems Timothy R. Radabaugh, Adriana Sampayo-Reyes,† Robert A. Zakharyan, and H. Vasken Aposhian* Department of Molecular and Cellular Biology, The University of Arizona, Tucson, Arizona 85721-0106 Received December 28, 2001
An arsenate reductase has been partially purified from human liver using ion exchange, molecular exclusion, hydroxyapatite chromatography, preparative isoelectric focusing, and electrophoresis. When SDS-β-mercaptoethanol-PAGE was performed on the most purified fraction, two bands were obtained. One of these bands was a 34 kDa protein. Each band was excised from the gel and sequenced by LC-MS/MS, and sequest analyses were performed against the OWL database SWISS-PROT with PIR. Mass spectra analysis matched the 34 kDa protein of interest with human purine nucleoside phosphorylase (PNP). The peptide fragments equal to 40.1% of the total protein were 100% identical to the corresponding regions of the human purine nucleoside phosphorylase. Reduction of arsenate in the purine nucleoside arsenolysis reaction required both PNP and dihydrolipoic acid (DHLP). The PNP rate of reduction of arsenate with the reducing agents GSH or ascorbic acid was negligible compared to that with the naturally occurring dithiol DHLP and synthetic dithiols such as BAL (British anti-lewisite), DMPS (2,3-dimercapto-1-propanesulfonate), or DTT (R-dithiothreitol). The arsenite production reaction of thymidine phosphorylase had approximately 5% of such PNP activity. Phosphorylase b was inactive. Monomethylarsonate (MMAV) was not reduced by PNP. The experimental results indicate PNP is an important route for the reduction of arsenate to arsenite in mammalian systems.
Introduction Inorganic arsenate and arsenite are present in carcinogenic concentrations in the drinking water of more than 2 million people throughout the world (1-6). Concerns about this problem have led to major advances in understanding the biochemistry and toxicology of inorganic arsenic (7-18). They have involved primarily the enzymology of arsenite methylation and its methylated species in mammalian systems. These methylations were until recently considered to be detoxification reactions, but are now considered to be bioactivation, since intermediates such as monomethylarsonous acid [MMA(III)]1 and dimethylarsenous acid [DMA(III)] have recently been demonstrated to be more toxic than arsenate or arsenite in vivo in hamsters (19), in vitro in human hepatocytes (20), and in other cell culture systems (21). Yet, very little is understood about the enzyme(s) responsible for the reduction of arsenate to the more toxic * To whom correspondence should be addressed at the Department of Molecular and Cellular Biology, 1007 E. Lowell, Life Sciences South Building, Room 444, P.O. Box 210106, The University of Arizona, Tucson, AZ 85721-0106. Phone: 520-621-7565, Fax: 520-621-3709, Email:
[email protected]. † Permanent address: Centro de Investigaciones Biomedicas del Noreste, I.M.S.S., Monterrey, Nuevo Leon Mexico. 1 Abbreviations: MMA(III), monomethylarsonous acid; DMA(III), dimethylarsenous acid; PNP, purine nucleoside phosphorylase; DTT, dithiothreitol; GSH, glutathione; BAL, 2,3-dimercaptopropanol; DMPS, sodium 2,3-dimercapto-1-propanesulfonate; MOPS, 3-(N-morpholino)propanesulfonic acid; DHLP/R-DHLP, dihydrolipoic acid.
arsenite or the oxidation of arsenite to arsenate in mammalian systems. We have previously reported an enzyme activity from human liver that reduced arsenate to arsenite (22). The enzyme, when purified 24-fold, required for activity a heat-stable cofactor less than 3 kDa in size plus a thiol. The heat-stable cofactor did not appear to be Cu2+, Mn2+, Zn2+, Mg2+, or Ca2+. The identification of this enzyme activity demonstrated that in humans the reduction of arsenate to arsenite is enzymatically catalyzed and is not solely, if at all, the result of a nonenzymatic chemical reduction by GSH as has been proposed in the past. In the present paper, we report that human liver arsenate reductase and human purine nucleoside phosphorylase (PNP) have amino acid sequence homology and that the heat-stable cofactor of arsenate reductase is inosine. We have used calf spleen PNP (23) to investigate the requirements of the reduction of arsenate to arsenite. These requirements include inosine and a dithiol for which DHLP is the most active naturally occurring dithiol.
Experimental Procedures Caution: Arsenic has been classified as a human carcinogen by the International Agency for Research on Cancer (24). Reagents. Sodium arsenate was ACS reagent grade and was purchased from MCB Reagents (Cincinnati, OH); carrier-free [73As]arsenate was from Los Alamos National Laboratory (Los
10.1021/tx0101853 CCC: $22.00 © 2002 American Chemical Society Published on Web 04/26/2002
Arsenate Reductase and PNP Alamos, NM); DEAE-Sephacel, Sephacryl S-200 HR, and molecular mass standards were from Pharmacia Biotech (Uppsala, Sweden). From Sigma Chemical Co. (St. Louis, MO) were purchased PMSF (phenylmethylsulfonyl fluoride), inosine, EDTA (ethylenediaminetetraacetic acid), sodium m-arsenite, hypoxanthine, 5′AMP, DTT (R-dithiothreitol), GSH (glutathione reduced form), BAL (2,3-dimercaptopropanol), ascorbic acid, β-mercaptoethanol, and lipoic acid (DL-6,8-thioctic acid reduced form, DHLP). A stock solution of DHLP was prepared in absolute ethanol, and the final concentration of ethanol in incubation mixtures was 1% or less. Calf spleen PNP (purine nucleoside phosphorylase), 2 units/ mg with bovine serum albumin as a stabilizer; rabbit muscle phosphorylase b, 21 units/mg; Escherichia coli thymidine phosphorylase, EC 2.4.2.4, 10.7 units/mg with bovine serum albumin as a stabilizer; thioredoxin from spirulina; and glycogen type IX from bovine liver were purchased from Sigma Chemical Co. Cow milk xanthine oxidase, 20 units/mL, was from Roche. DMPS (sodium 2,3-dimercapto-1-propanesulfonate) was from Heyl (Berlin, Germany). Monoflow III scintillation fluid was obtained from National Diagnostics (Atlanta, GA). All other reagents were analytical reagents or of the highest quality available. Tissue. Human liver from a male was obtained from the Association of Human Tissue Users (Tucson, AZ). Partial Purification of Arsenate Reductase. (A) Cytosol Preparation. Human liver (225 g) was divided into 5 g pieces, minced with scissors, and homogenized with 10 mL of homogenization buffer in a Dounce homogenizer. Homogenization buffer contained 20 mM Tris-HCl (pH 8.0), 5 mM EDTA, 1 mM DTT, and 0.2 mM PMSF. The homogenate was centrifuged at 12000g for 15 min at 4 °C. Lipids were skimmed from the supernatant by spatula, and the supernatant was then centrifuged at 105000g for 90 min at 4 °C. Lipids were removed from the supernatant by filtration through glass wool, and the resulting cytosol was stored at -70 °C. (B) DEAE Chromatography. Cytosol (250 mL containing 8350 mg of protein) was thawed and diluted with 500 mL of 10 mM Tris-HCl buffer, pH 7.5, and loaded onto a 2.5 × 18 cm DEAE Sepharose Fast Flow (Amersham Pharmacia Biotech) column equilibrated with 30 mM Tris-HCl, pH 7.5. The sample was loaded at 2 mL/min, and the column was washed with 1 column volume of equilibration buffer. Elution was carried out with a gradient of 250 mL of 30 mM Tris-HCl (pH 7.5) and 250 mL of 30 mM Tris-HCl/300 mM NaCl (pH 7.5) at a flow rate of 300 mL/h. Fractions containing the highest arsenate reductase activity were pooled and concentrated by ultrafiltration using a 30 kDa cutoff membrane. Fractions were not stable unless concentrated before freezing. The concentrate was stored at -70 °C. A portion of the thawed concentrate, 87.5 mg of protein, was transferred into 10 mM potassium phosphate buffer (pH 6.8) using PD-10 desalting columns (Amersham Pharmacia Biotech). (C) Hydroxyapatite Chromatography. The protein was then applied to a 1.5 × 17 cm column of 80 µm Type I ceramic hydroxyapatite (BioRad Laboratories) and equilibrated with 10 mM potassium phosphate buffer (pH 6.8). The column was washed with 2 column volumes of the equilibration buffer and then eluted with a continuous gradient of 150 mL of 10 mM potassium phosphate (pH 6.8) and 240 mM potassium phosphate (pH 6.8) at a flow rate of 2 mL/min. Collected fractions were transferred into 30 mM Tris-HCl buffer, pH 7.35, using PD-10 columns and then assayed. Fractions containing the highest activity were pooled, concentrated, and stored at -70 °C. (D) Sephacryl S-200 Chromatography. A portion of the concentrated protein (52.5 mg) from the hydroxyapatite column was applied to a 2.5 × 62 cm Sephacryl S-200 HR column equilibrated with 30 mM Tris-HCl/50 mM NaCl, pH 7.5. Active fractions were pooled, concentrated, and stored at -70 °C. (E) Preparative Isoelectric Focusing. Concentrated protein (280 µL and 3.44 mg) from the Sephacryl column was loaded
Chem. Res. Toxicol., Vol. 15, No. 5, 2002 693 onto a 60 mL Rotofor Cell (BioRad Laboratories). The cell was prefocused with Bio-Lyte ampholytes pH 3-10 (BioRad Laboratories) according to the manufacturer’s instructions. After loading, the sample was focused for 2.5 h at 12 W constant power. Fractions were collected (some fractions required centrifugation to remove precipitated proteins) and assayed. Active fractions were pooled, concentrated, and stored at 4 °C overnight. (F) Preparative Electrophoresis. Concentrated protein, 1.8 mg, in 20% buffered glycerol, pH 7.35, from the Rotofor Cell was applied to a Mini Prep Cell (BioRad Laboratories) for preparative electrophoresis. Electrophoresis using a 6% native PAGE column gel was carried out at 100 V, 4 °C, using 30 mM Tris-HCl buffer (pH 7.35) for the elution buffer at a flow rate of 100 µL/min. Active fractions were combined into seven pools (designated A, B, C, D, E, F, and G) for further analysis and stored at -70 °C. Mass Spectrometry of Partially Purified Arsenate Reductase. Electrophoresis of equal amounts of protein (2 µg) from samples A, B, C, D, E, F, and G was performed using a 7.5% SDS-polyacrylamide gel and then stained with Coomassie blue. The gel was photographed wet, and three major bands were excised and delivered to the Analytical Core Facility of the Southwest Environmental Health Science Center at The University of Arizona for analysis. The gel slices were digested with trypsin (25), and the extracted peptides were analyzed by liquid chromatography-tandem mass spectrometry (LC-MS/MS) using a quadrupole ion trap Finnigan LCQ Classic mass spectrometer equipped with a Finnigan MAT Spectra System quaternary pump P4000 HPLC and a Finnigan electrospray ion source (Thermoquest, San Jose, CA). Peptides were eluted from a reversed-phase C18 micro-column (Vydac 250 mm × 1 mm, Hesperia, CA) with a gradient of 3-95% acetonitrile in 0.5% formic acid, 0.01% trifluoroacetic acid over 150 min at a flow rate of 15 µL/min. Tandem MS spectra of peptides were analyzed with the Sequest program (v. C1) to assign peptide sequences to the spectra (25). Sequest analyses were performed against the OWL database SWISS-PROT with PIR. Arsenate Reductase Assay Used during Enzyme Purification Procedures. The assay used during purification was the same as described previously (22). Arsenite and arsenate were separated on a 4.9 × 250 mm Hamilton PRP X-100 anion exchange HPLC column using 50 mM sodium phosphate buffer (pH 8.0) with a flow rate of 2.5 mL/min. Arsenicals were detected and quantified with a postcolumn inline β-RAM radioisotope detector (INUS Systems Inc., Tampa, FL). Monoflow III (National Diagnostics, Atlanta, GA) scintillation cocktail was used at a flow rate of 7.5 mL/min. Arsenate Reductase Assay Using PNP/HPLC. The reaction mixture (100 µL) containing 0.1 M MOPS buffer, pH 7.5, 1.0 mM EDTA, 10 mM inosine, 0.37-6 mM arsenate, 0.5 µCi of carrier-free [73As]arsenate, 0.5-5 mM DHLP, and PNP or human liver arsenate reductase was incubated for 30 min at 37 °C. Arsenite, 0.1 M (100 µL) after incubation, was added rapidly, and the tubes were immersed in a boiling water bath for 2 min, cooled on ice for 5 min, and then centrifuged at 12 000 rpm for 6 min. A 100 µL sample of the supernatant was injected onto a 4.9 mm × 250 mm Hamilton PRP-X100 anion exchange column using 30 mM sodium phosphate (pH 5.0) with a flow rate of 1.45 mL min-1. The arsenite and arsenate were quantified using a postcolumn inline Beckman 171 radioisotope detector. Monoflow III was used as the scintillation fluid, and a flow rate was used that was 3 times that of the mobile phase. Purine Nucleoside Phosphorylase (PNP), Spectrophotometric Assay. PNP activity with inosine as the substrate was measured by a spectrophotometric assay, which monitors the change in A293 associated with the formation of uric acid by xanthine oxidase. A molar absorption coefficient of 1.25 × 104 was used (26). The assay was performed in a final volume of 1 ml by incubating 0.5 mM inosine, 5 mM, 1 µg of PNP, and 0.02 unit of xanthine oxidase in 50 mM potassium phosphate buffer, pH 7.5 at 30 °C, for 5 min (26).
694
Chem. Res. Toxicol., Vol. 15, No. 5, 2002
Radabaugh et al.
Figure 1. (A) SDS-β-mercaptoethanol-polyacrylamide gel analysis of the fraction from preparative electrophoresis. Each lane contained 2 µg of protein of fractions A-G. (B) Arsenate reductase activity of fractions A-G from the nondenaturing preparative electrophoresis column (see Experimental Procedures). The PNP arsenolysis activity with inosine as the substrate was performed as above except potassium phosphate buffer was replaced with 100 mM MOPS buffer, pH 7.5, and 20 mM arsenate and incubated at 30 or 37 °C. Thymidine Phosphorylase, HPLC Assay. The assay (100 µL) was performed as above for PNP, only MOPS buffer was replaced with 20 mM Tris, pH 7.3, in the presence of 6 mM arsenate (pH 7.3), 10 mM thymidine, 1-5 mM DHLP, and 11 µg of thymidine phosphorylase at 37 °C for 30 min. The arsenite and arsenate were measured as above. Thymidine Phosphorylase, Spectrophotometric Assay. The thymidine phosphorylase activity with thymidine as the substrate was measured by a spectrophotometric assay, which monitors the change in A300 associated with formation of thymine. The assay was performed in a final volume of 1 mL by incubating 5 mM thymidine, 10 mM arsenate in Tris buffer, 10 mM, pH 7.3, with 5.5 or 11 µg of thymidine phosphorylase at 37 °C for 5 min as described previously (27). Phosphorylase b Assay. The assay (100 µL) was performed with 230 µg of glycogen, 6 mM arsenate, 2 mM EDTA (pH 7.5), 10 mM MgCl2, 5 mM DHLP, 0.5 µCi, of carrier-free 73As, 1 mM 5′AMP in 50 mM Tris-acetate, pH 7.5, 10 µg of rabbit muscle phosphorylase b. The reaction was started by addition of 5′AMP (28) and incubated for 30 min at 37 °C. The arsenite and arsenate were measured as above using HPLC. Other Methods. The protein concentration was determined by the method of Bradford using BSA as a standard (29).
Results Partial Purification of Human Liver Arsenate Reductase. Fractions D and E (Figure 1B) of the last purification step had a specific activity of 1705 nmol of arsenite (mg of protein)-1 (30 min)-1 compared with 2.8 nmol of arsenite (mg of protein)-1 (30 min)-1 in the cytosol resulting in a 609-fold purification. Fractions D and E contained 250 µg of protein and 7.3% of the initial activity of 2088 mg of protein. (Protein was normalized to reflect only protein carried through to the last step of purification. Significant amounts of purified protein were set aside for other purposes after the DEAE and Sephacryl steps.) Fractions D and E contained two major bands (Figure 1A). One of these bands is a ∼34 kDa protein (Figure 1A). Mass Spectrometry Analysis of Partially Purified Human Liver Arsenate Reductase. Mass spectra analysis using Sequest in conjunction with the SWISS-
Figure 2. Sequence identity between human purine nucleoside phosphorylase (PNP) and partially purified human liver arsenate reductase. The sequence of PNP is aligned with arsenate reductase peptide fragments determined by mass spectrometry. PNP is row 1; human liver arsenate reductase is row 2. Table 1. Dihydrolipoic Acid Is Required for the Reduction of Arsenate by PNPa arsenite produced, nmol assay-1 (30 min)-1 ( SE complete reaction mixture minus PNP minus DHLP
103.6 ( 2.9 13.6 ( 1.0 2.09 ( 0.11
a The complete reaction mixture (100 µL) contained inosine 10 mM, arsenate 6 mM, DHLP 5 mM, EDTA 1 mM in MOPS buffer, pH 7.5, 1 µg of PNP, 30 min at 30 °C, n ) 3.
PROT database matched the ∼34 kDa protein of interest (Figure 1A) with human purine nucleoside phosphorylase (accession no. AAA36460). The peptide fragments equal to 40.1% of the total protein were 100% identical to the corresponding regions of the human purine nucleoside phosphorylase (Figure 2). Dihydrolipoic Acid Is Required by Arsenate Reductase. Since we could not obtain human PNP, we used calf spleen PNP for these studies. The homology of the calf spleen enzyme with the human enzyme is 87% overall and 100% within the active site (30, 31). Reduction of arsenate in the purine nucleoside arsenolysis required both PNP and DHLP (Table 1). In the absence of DHLP, the amount of arsenite formed was decreased to about 2% of that found with the complete reaction. If PNP was omitted, the reduction of arsenate was 13% of that found with the complete reaction. Significant activity was observed only in the presence of PNP and DHLP. The reduction of arsenate coupled with the arsenolysis of inosine was linear to at least 1.0 µg. This amount, therefore, was used for the remainder of experiments since it was in the linear range of the assay. Substrate saturation for the arsenate reductase (PNP) occurred at about 6 mM arsenate. The rate of arsenite production depended on the concentration of DHLP; Michaelis-Menten kinetics indicated that the arsenate reductase activity of PNP in the presence of 2 and 5 mM DHLP had Km ) 0.81 × 10-3 M, Vmax ) 62.5 µmol mg-1
Arsenate Reductase and PNP
Chem. Res. Toxicol., Vol. 15, No. 5, 2002 695 Table 2. Dihydrolipoic Acid Did Not Influence Hypoxanthine Production during Phosphorolysis or Arsenolysis Using PNP (n ) 4) hypoxanthine formed, µmol mg-1 min-1 ( SE phosphorolysisa minus DHLP 5 mM DHLP
arsenolysisb
30 °C
30 °C
37 °C
2.4 ( 0.16 2.6 ( 0.04
4.42 ( 0.06 4.16 ( 0.29
6.78 7.31 ( 0.116
a Purine nucleoside phosphorylase activity with inosine as the substrate was measured by the spectrophotometric assay, which monitors the change in A293 associated with the formation of uric acid by xanthine oxidase. A molar absorption coefficient of 1.25 × 104 (Stoeckler, 26) was used. The assay was performed by incubating at 30 °C for 5 min inosine 0.5 mM, DHLP 5 mM, PNP 1 µg, and xanthine oxidase 0.02 unit in 50 mM potassium phosphate buffer, pH 7.5. The final reaction volume was 1 mL. b Purine nucleoside arsenolysis activity with inosine was performed as above except potassium phosphate buffer was replaced with 100 mM MOPS buffer, pH 7.5, and 20 mM arsenate.
Figure 3. Dihydrolipoic acid and the reduction of arsenate by PNP. The assay (100 µL) was performed using inosine 10 mM, arsenate 6 mM, EDTA 1 mM, and PNP 1 µg in 0.1 M MOPS buffer, pH 7.5, 30 min at 37 °C.
Figure 5. Arsenate reductase activity of PNP with different thiols. The assay (100 µL) was performed using 10 mM inosine, 6 mM arsenate, 0.5-5 mM DHLP or GSH, DTT, BAL, DMPS, ascorbic acid, β-metcaptoethanol, and PNP (1 µg) in 0.1 M MOPS buffer, pH 7.5, 30 min at 37 °C.
Figure 4. Michaelis-Menten kinetics for PNP arsenolysis and reduction of arsenate. DHLP was present at 5 or 2 mM, 1 µg of PNP. For 5 mM DHLP, Km ) 1.8 × 10-3 M and Vmax ) 166 µmol mg-1 (30 min)-1. For 2 mM DHLP, Km ) 0.81 × 10-3 M and Vmax ) 62.5 µmol mg-1 (30 min)-1. S ) concentration of arsenate (see Experimental Procedures).
(30 Km ) 1.8 × M, Vmax ) 166 µmol (30 min)-1, respectively (Figure 4). Hypoxanthine production by either PNP phosphorolysis or arsenolysis of inosine (Table 2) was not influenced by DHLP. Monomethylarsonate (MMAV) was not reduced by PNP (data not shown). Among different thiols tested in the reduction of arsenate to arsenite by the arsenolysis reaction (Figure 5), the most active naturally occurring thiol was DHLP. The rate of reduction of arsenate by reducing agents such as GSH or ascorbic acid was negligible compared to that of the endogenous dithiol DHLP and the synthetic DTT (Figures 3 and 5). min)-1;
10-3
mg-1
Michaelis-Menten kinetics for PNP in the arsenolysis and reduction of arsenate using increasing concentrations of DHLP yielded a Km ) 2.6 × 10-3 M and Vmax ) 143 µmol mg-1 (30 min)-1 (Figure 6). Arsenate and Thymidine Phosphorylase, PNP, or Phosphorylase b. To test the uniqueness of PNP for arsenate reduction, other phosphorylases were investigated. At different concentrations of DHLP (1, 2.5, or 5 mM), the arsenolysis reaction of thymidine phosphorylase had approximately 5% of the PNP arsenolysis activity (Table 3). Rabbit muscle phosphorylase b was inactive in the reduction of arsenate to arsenite.
Discussion Arsenate Reductase and PNP. The purification and characterization of the enzymes involved in the biotransformation of inorganic arsenic are crucial for understanding the mechanism of inorganic arsenate and arsenite metabolism. We have determined some of the amino acid sequences of partially purified human liver arsenate reductase. Based on amino acid homology, our evidence indicates that it is identical to human purine nucleoside phosphorylase (PNP). PNP is an essential enzyme of purine metabolism. Many sophisticated biochemical studies of it are available [for a review, see Bzowska et al. (32)]. It is specific for purine and purine analogues, ribonucleosides, and des-
696
Chem. Res. Toxicol., Vol. 15, No. 5, 2002
Radabaugh et al.
Figure 6. Michaelis-Menten kinetics for PNP arsenolysis and reduction of arsenate. S ) Different concentrations of DHLP; arsenate concentration was 6 mM and PNP concentration was 1 µg. Vmax ) 143 µmol mg-1 (30 min)-1; Km ) 2.6 × 10-3 M. Table 3. Arsenate and Thymidine Phosphorylase, PNP, or Phosphorylase ba arsenite produced, nmol DHLP (mM)
thymidine phosphorylase
PNP
phosphorylase b
1 2.5 5
2.2 3.5 4.9
37.3 85.6 101.4 ( 3.08
